RETAIN data highlight role of ctDNA in bladder preservation for MIBC

The phase 2 RETAIN trial (NCT02710734) met its primary end point of metastasis-free survival (MFS), and circulating tumor DNA (ctDNA) data indicated the biomarker has strong utility as a “prognostic marker for systemic disease control,” Pooja Ghatalia, MD, reported at the 2026 American Society of Clinical Oncology Genitourinary Cancers Symposium.¹

“Historically, the standard of care for muscle-invasive bladder cancer [MIBC] has been neoadjuvant therapy followed by cystectomy or trimodal therapy. But we know cystectomy is life-altering and significantly impacts quality of life [for] our patients. Importantly, 35% to 60% of patients achieve a pathologic complete response after neoadjuvant therapy, creating a clear opportunity for response-adapted bladder preservation trials,” explained Ghatalia, an associate professor of hematology/oncology and a collaborating member of Cancer Epigenetics in the Nuclear Dynamics and Cancer Research Program at Fox Chase Cancer Center in Philadelphia, Pennsylvania.

For the single-arm phase 2 RETAIN-2 trial, patients were eligible for inclusion if they had cT2-T3N0 disease, predominant urothelial carcinoma of the bladder, and an ECOG performance status of 0 or 1. Pretreatment tissue from patients was analyzed for mutational biomarkers that linked with response to cisplatin treatment. Following transurethral resection of bladder tumor (TURBT), patients received 3 cycles each of neoadjuvant accelerated methotrexate, vinblastine, doxorubicin, and cisplatin plus nivolumab (Opdivo). Following this treatment, patients received another TURBT, along with a CT scan, urine cytology, and biopsy. If patients were mutation positive and had no residual disease, they underwent active surveillance. If they had no mutation, no residual disease, or a positive urine cytology, they underwent treatment consisting of either intravesical treatment, chemotherapy/radiation therapy, or cystectomy (determined via shared decision-making).

The primary end point was 2-year MFS, defined as absence of surgically unresectable local recurrence, nodal recurrence, and M1 disease. Follow-up was 5 years. The primary object of the study was to determine whether the risk-adapted strategy utilized for the study was noninferior to standard-of-care neoadjuvant chemotherapy followed by cystectomy for all patients.

A total of 80 patients began the study, of whom 9 were not evaluable. Treatment was recommended in 47 patients, and active surveillance was recommended for 23 patients. One patient was lost to follow-up after the second TURBT procedure. Following crossover between the cohorts, 48 patients received treatment and 22 underwent active surveillance.

Discussing baseline patient characteristics, Ghatalia noted that 32% of patients had urothelial carcinoma/variant histology, 42% had cT3 disease, and 44% were mutation positive.

At a median follow-up of 32.3 months, 70% of patients in the intent-to-treat (ITT) population were metastasis free at 2 years.

“The lower bound of the 1-sided 90% confidence interval was 62.4%, well exceeding our noninferiority threshold of 56%, so RETAIN-2 met its primary end point,” Ghatalia said.

The 2-year MFS rate was 83.5% (95% CI, 75-92.9) and 85.2% (95% CI, 70.9-100) for the ITT population and the active surveillance cohort, respectively.

For patients who received treatment, 35 underwent cystectomy. In this group, the ypT0 rate was 46%. Eight patients who underwent cystectomy had metastases. Ten patients underwent chemo/radiation; 1 of these patients had metastases. Three patients received treatment with intravesical Bacillus Calmette-Guérin; 1 of these patients underwent salvage cystectomy.

Ghatalia then examined the active surveillance cohort. Fourteen of these patients were recurrence free, 4 had recurrence of non-MIBC, 3 had recurrence of MIBC, and 1 had pelvic nodal recurrence.

“Overall, 68% of 22 patients on active surveillance remained metastasis free with an intact, nonirradiated bladder. All metastatic events were preceded by local recurrence,” Ghatalia said.

Comparing RETAIN-1² with RETAIN-2, Ghatalia noted that 62% of patients in the active surveillance cohort of RETAIN-1 had local recurrence and 36% had metastases compared with 36% and 14% of patients undergoing active surveillance in RETAIN-2, respectively. In addition, 48% of patients undergoing active surveillance in RETAIN-1 were metastasis free with an intact, nonirradiated bladder compared with 68% of patients undergoing active surveillance in RETAIN-2.

Ghatalia then discussed the ctDNA analysis in RETAIN-2. Between RETAIN-1 and RETAIN-2, she said, the investigators analyzed 275 time points from 111 patients. The investigators utilized the Signatera test for the analysis. For both trials, data were available at baseline and postneoadjuvant treatment; for RETAIN-2, data were also available at 3 months and 6 months.

Among the 111 patients, 50 (45%) underwent cystectomy, 42 (38%) underwent active surveillance, 12 (11%) received chemoradiation, and 7 (6%) underwent intravesical therapy. Baseline ctDNA positivity was 42.3%, and posttreatment ctDNA positivity was 13.6%. The overall rate of ctDNA clearance was 72.4% (32 of 44 patients). In patients undergoing active surveillance, the ctDNA clearance rate was 83.3% (15 of 18 patients); for patients undergoing treatment, the clearance rate was 65.4% (17 of 26 patients).

“This clearance correlated strongly with outcomes. Both baseline…and postneoadjuvant…ctDNA positivity were strongly associated with inferior [MFS and] overall survival,” Ghatalia said.

She added, “When we looked at ctDNA dynamics, patients who had baseline and postneoadjuvant ctDNA negativity had the best outcomes. Those who cleared ctDNA…did a little bit worse, and those who remained persistently ctDNA positive had the worst outcomes. In RETAIN-2 patients, where serial postneoadjuvant time points were available, this pattern was confirmed. Any ctDNA positivity correlated with worse outcomes.”

The investigators then considered whether ctDNA might be able to aid treatment selection.

“When we stratified [MFS] by treatment type—cystectomy vs active surveillance—and postneoadjuvant therapy ctDNA status, a clear picture emerged. ctDNA-positive patients had poor outcomes despite undergoing cystectomy, suggesting that this group may need additional systemic therapy or that less radical local therapy could be offered,” Ghatalia said.

When examining ctDNA related to pathologic staging, Ghatalia reported, “Among cystectomy patients, postneoadjuvant therapy ctDNA status provided prognostic information beyond pathologic staging.

“RETAIN-2 met its primary end point. Sixty-eight percent of active surveillance patients were metastasis free with an intact, nonirradiated bladder at this 32-month follow-up. ctDNA is an extremely prognostic marker for systemic disease control. ctDNA positivity [is] associated with a 10.7-fold higher risk of inferior [MFS and]…poor outcomes despite cystectomy. ctDNA does not reliably capture residual bladder disease or local recurrence. Future trials must integrate ctDNA with clinical restaging and urinary biomarkers,” Ghatalia concluded.

REFERENCES

1. Ghatalia P, Ross, EA, Zibelman MR, et al. Circulating tumor DNA (ctDNA) to guide response-adapted bladder preservation in muscle invasive bladder cancer (MIBC): integrated analysis of the RETAIN trials. J Clin Oncol. 2026;44(suppl 7):LBA632. doi:10.1200/JCO.2026.44.7_suppl.LBA632

2. Geynisman DM, Abbosh PH, Ross E, et al. Phase II trial of risk-enabled therapy after neoadjuvant chemotherapy for muscle-invasive bladder cancer (RETAIN 1). J Clin Oncol. 2025;43(9):1113-1122. doi:10.1200/JCO-24-01214